Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

101 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SoilGuard Portal: Soil Quality & Agricultural Monitor

Privacy No data sold NZ Privacy Act Te Mana Raraunga NZ AI Safety SOC2 Regions Security Governance

Coastal Alpine Tech portfolio

Stage Hybrid HITL Te Mana Raraunga

Part of the Kiwi Edge AI Stack | Founder OS: NZ-Start-Up | Agent policy: .github/agent-fleet/

Sovereign hybrid edge AI for NZ farms and founders - local-first + multi-model, Te Mana Raraunga aligned - collaborating with Venture Taranaki, startups.com investors and Kotahitanga Investment Fund (HITL + cultural advisory for formal approaches).

Agents inform, draft, prepare, monitor, and remind. Humans advise, sign, file, send, and pay. Anti-hallucination policy: .github/agent-fleet/anti-hallucination.md | Congruence: CAT_CONGRUENCE.md

Privacy / Security / Governance

Coastal Alpine Tech products treat operational and personal data as taonga. Defaults favour local-first operation, purpose-limited collection, and Human-in-the-Loop for high-stakes actions.

Hard commitments

Commitment Statement
No data sales We do not sell personal information or customer operational data to third parties for advertising, brokerage, or unrelated commercial exploitation.
NZ Privacy Act 2020 Collection, use, storage, and disclosure of personal information is designed to operate in accordance with the Privacy Act 2020 information privacy principles (including IPP awareness and IPP 3A indirect-collection notification where applicable).
Te Mana Raraunga Where Māori data or community data interests arise, systems are designed to operate in accordance with Te Mana Raraunga principles (including Rangatiratanga, Whakapapa, Whanaungatanga, Kotahitanga, Manaakitanga, Kaitiakitanga) as a sovereignty and stewardship lens — not as a marketing slogan.
NZ AI safety AI features follow a NZ AI safety-aligned posture: Algorithm Charter spirit (fairness, transparency, human oversight where relevant), digital.govt.nz / responsible AI guidance awareness, no silent model training on private journals without consent, and HITL for high-stakes outcomes.
Security No silent exfiltration; owner-controlled credentials; least privilege; SecOps / dependency hygiene on the fleet cadence.
Governance Agents inform, draft, prepare; humans advise, sign, file, send, and pay.
Pillar Commitment
Privacy Local-first / offline-capable where practical; Privacy Act 2020; Te Mana Raraunga spirit; third-party AI only when opt-in and labelled
Security No silent exfil of tenant or personal data; owner-controlled keys
Governance HITL for high-stakes; Te Mana Raraunga spirit; multi-region compliance maps in COMPLIANCE_REGIONS.md

Agents inform, draft, prepare, monitor, and remind. Humans advise, sign, file, send, and pay.

Fleet policy: fivepanelhat / Kiwi Edge AI Stack · COMPLIANCE.md · COMPLIANCE_REGIONS.md · SECURITY.md

Problems we are solving

SoilGuard targets soil moisture, nutrients, and nitrate-cap discipline for NZ whenua and pasture systems.

  1. Rural blackouts - Irrigation and compliance logs must not depend on cloud uptime.
  2. Nitrate caps - NES-F synthetic N limits create real fine risk without local evidence trails.
  3. Runoff and erosion - Over-irrigation and poor timing damage waterways and consents.
  4. Customary data rights - Whenua telemetry belongs under local custody (Te Mana Raraunga).

Solution we have built

Built capability What it does
Multi-modal soil ingestion Moisture, EC, N-P-K, vision, acoustic watchdogs
Local inference Ollama / Gemma on RPi 5 + Hailo path
Actuation hooks Irrigation / fertigation under HITL allow-lists
Council-ready ledgers Export patterns for regional reporting

Local (Taranaki) and national (Aotearoa) economic benefits

Lever Benefit
Regional R&D HQ Product design and IP stay in New Plymouth / Taranaki - not only Auckland/offshore SaaS
Primary-sector productivity On-farm and rural tools aim to cut waste, protect consents, and support export competitiveness
Skilled employment pathways Edge install, field support, agritech ops, software, compliance, and cultural advisory roles as pilots scale
Data sovereignty Te Mana Raraunga-aligned local custody keeps high-value operational data onshore
HITL jobs quality Agents inform / draft / prepare / monitor / remind; humans advise / sign / file / send / pay - augment people, do not fake full autonomy

Stage honesty (pre-seed): Impact today is founder R&D, near-term contractors, and EDA/partner leverage. Permanent multi-region payroll follows paid pilots and revenue - we do not invent headcount claims.

License: Proprietary Python Ollama

Linux Windows macOS Raspberry Pi

Claude AI Gemini OpenAI Grok

Hailo NPU MQTT Data Sovereign Sustainability

CI Status SecOps RedTeam Dependencies

SoilGuard Portal Banner

Coastal Alpine Tech Limited pre-seed startup, New Plymouth, Taranaki, Aotearoa New Zealand. Edge AI | Sovereign Systems | Practical Intelligence

Autonomous on-premise soil quality monitoring and agricultural control system for New Zealand dairy farms, glasshouses, and orchards. Powered by local edge AI, it runs fully offline in remote and regulated rural catchments to maintain data sovereignty.


Architecture Overview

Diagrams: Architecture images and Mermaid maps describe the target product architecture for this pre-seed stack. They are engineering design maps not claims of large-scale commercial fleet deployment.

SoilGuard monitors soil health and fertigation on-device. Sensor streams and optional vision feed Gemma 4 via Ollama on RPi 5 16GB + Hailo-10H, with actuator lockouts for regulatory safety.

SoilGuard architecture liquid glass overview

System map

%%{init: {
 "theme": "dark",
 "themeVariables": {
 "fontSize": "16px",
 "fontFamily": "Inter, ui-sans-serif, system-ui, sans-serif",
 "primaryColor": "#0ea5e9",
 "primaryTextColor": "#f8fafc",
 "primaryBorderColor": "#38bdf8",
 "lineColor": "#67e8f9",
 "secondaryColor": "#1e293b",
 "tertiaryColor": "#0f172a",
 "clusterBkg": "#0b1220cc",
 "clusterBorder": "#38bdf880",
 "titleColor": "#e2e8f0"
 },
 "flowchart": {
 "nodeSpacing": 40,
 "rankSpacing": 48,
 "padding": 20,
 "htmlLabels": true,
 "curve": "basis"
 }
}}%%
flowchart TB

 classDef sense fill:#052e16,stroke:#4ade80,stroke-width:2px,color:#f0fdf4
 classDef edge fill:#0c4a6e,stroke:#38bdf8,stroke-width:2px,color:#f0f9ff
 classDef core fill:#134e4a,stroke:#2dd4bf,stroke-width:2px,color:#f0fdfa
 classDef act fill:#422006,stroke:#fbbf24,stroke-width:2px,color:#fffbeb
 classDef store fill:#1e1b4b,stroke:#a5b4fc,stroke-width:2px,color:#eef2ff
 classDef ai fill:#3b0764,stroke:#e879f9,stroke-width:2px,color:#fdf4ff
 classDef app fill:#1e1b4b,stroke:#c4b5fd,stroke-width:2px,color:#eef2ff

 subgraph IN[" Field inputs"]
 S["Soil probes<br/>N-P-K | moisture | pH | EC"]
 V["Vision optional<br/>leaf / canopy cues"]
 end

 subgraph EDGE[" Edge agent RPi 5 16GB + Hailo-10H"]
 CORE["Coastal-Alpine-Core"]
 LLM["Gemma 4 via Ollama"]
 AG["SoilGuard AI agent"]
 end

 subgraph OUT[" Control & records"]
 FERT["Fertigation lockouts / setpoints"]
 ALERT["Operator alerts"]
 STORE["Local telemetry | compliance logs"]
 end

 S & V --> CORE --> LLM --> AG
 AG --> FERT & ALERT & STORE

 class S,V sense
 class CORE,AG core
 class LLM ai
 class FERT,ALERT act
 class STORE store
Loading
Layer Components Role
Sensors N-P-K moisture
Agent Gemma 4 + Core SDK Offline decisions
Safety Fertigation lockouts NES-F / regional rules
Hardware RPi 5 16GB + Hailo-10H Canonical edge target

Full detail: ARCHITECTURE.md

The 5 Ws: Project Context

  • Who: Developed by Coastal Alpine Tech Limited intended for collaboration with NZ crop growers, dairy farmers, and iwi trusts (partnerships in development pre-seed).
  • What: An edge-native IoT monitor and agentic control system that ingests soil telemetry (moisture, temperature, EC, N-P-K), visual leaf health cues, and acoustic diagnostics to optimize crop yield and generate council-ready environmental audits.
  • Where: Deployed on-premise at nurseries, glasshouses, orchards, and pastoral runoff sites across New Zealand. HQ in New Plymouth, Taranaki.
  • When: Active development as of June 2026.
  • Why: Synthetic nitrogen fertilizer caps and Freshwater Farm Plan rules require farm operators to strictly manage soil runoffs and leaching. SoilGuard provides on-device data logging and localized automation without relying on vulnerable, non-sovereign cloud services.


Key Features

  • Multi-Modal Soil Ingestion: Real-time processing of moisture (%VWC), temperature, electrical conductivity, and dry-soil N-P-K nutrients via MQTT gateways.
  • Leaf Quality Inspection: Local USB/CSI camera frame capture processed via local Gemma 4 multimodal vision prompts to identify signs of crop chlorosis, pest incursions, or drying.
  • Acoustic Diagnostic Watchdog: Mic capture loop listening for irrigation equipment anomalies or valve blockages.
  • Google Gemma 4 (gemma4:e4b): Edge-quantized multimodal large language model executing offline on RPi 5 to assess conditions and issue control commands.
  • Automated Actuation: GPIO/PWM control loops for solenoid irrigation valves, nutrient fertigation pumps, and ventilation fans.
  • Council-Ready Audits: Instant export of CSV ledgers and JSON traces matching Waikato and Horizons regional reporting specifications.
  • Media Disk Protection: Active storage capacity management via media pruner that cleans old transient images/audio while preserving all compliance records.

Directory Structure

SoilGuard-Portal/
|-- portal_schemas/ # Pydantic schemas (compliance, readings, plans)
|-- portal_core/ # Core modules (config, mqtt, av, hardware, pruner)
|-- telemetry_data/ # Local ledger dumps and transient media buffers
|-- tests/ # Unit and security stress test files
|-- main.py # Unattended orchestrator entrypoint
|-- validate.py # diagnostics boot sequences
|-- setup.py # pip installation setup script
|-- soilguard.service # Systemd service unit template
|-- requirements.txt # Production package requirements
|-- requirements-dev.txt # Unit test requirements
|-- .env.example # Local configuration template
|-- ARCHITECTURE.md # Mermaid sequence flows and layout mapping
|-- COMPLIANCE.md # NZ Legislative mapping details
-- README.md # Project user documentation

Quick Start

Hardware Prerequisites

  • Raspberry Pi 5 (16GB RAM) Available locally via PB Tech or Kiwi Electronics.
  • Raspberry Pi AI Accelerator / AI HAT+ 2 (40 TOPS, Hailo-10H NPU) Key for offline generative AI workloads.
  • Soil Probes & ESP32 gateway Telemetry sensors broadcasting over MQTT.
  • USB/CSI Camera Installed above crop canopy in IP67 enclosure.

Installation & Setup

We provide separate guides for system environment setup and installation for Windows and Linux users:

Quick Start (Automated Setup)

The fastest way to install is running the cross-platform bootstrap script:

python bootstrap.py

d-Portal python bootstrap.py


### Manual Installation & Run

1. **Clone the Portal Repository:**

 ```bash
 git clone https://github.com/fivepanelhat/SoilGuard-Portal.git
 cd SoilGuard-Portal
  1. Configure Virtual Environment:
Linux / macOS (Bash)
python3 -m venv venv
source venv/bin/activate
pip install git+https://github.com/fivepanelhat/coastal-alpine-core.git@v0.2.0
pip install -r requirements.txt -r requirements-dev.txt
cp .env.example .env
Windows (PowerShell)
python -m venv venv
.\venv\Scripts\Activate.ps1
pip install git+https://github.com/fivepanelhat/coastal-alpine-core.git@v0.2.0
pip install -r requirements.txt -r requirements-dev.txt
Copy-Item .env.example .env

Note: If you receive an execution policy error, run Set-ExecutionPolicy -Scope CurrentUser RemoteSigned first.

  1. Deploy Gemma 4 Model:
ollama serve
ollama pull gemma4:e4b
  1. Verify System Setup:
python validate.py
  1. Start Orchestrator Daemon:
python main.py

Performance & Benchmarks

  • Local Inference Latency: ~0.85 seconds per query running Google's gemma4:e4b on Raspberry Pi 5.
  • Energy Consumption: Peak active NPU execution draw is ~1.5W, enabling solar-powered off-grid deployment.
  • Storage Footprint: media pruner limits raw camera frame buffer size below 500MB, retaining compliance records for 7+ years in compressed format.

Built for New Zealand data sovereign, edge-native, compliance-aware. Questions or collaboration? Contact Coastal Alpine Tech Limited, New Plymouth, Taranaki.

About

Soil & pasture domain portal — N-P-K / moisture edge architecture on Core SDK with SecurityGuard, flywheel, and RPi 5 16GB + Hailo-10H for NZ whenua.

Topics

Resources

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages